US20180336578A1 - Interest level evaluation method, interest level evaluating device and recording medium - Google Patents

Abstract

An interest level evaluation method executed by a processor of a computer, the interest level evaluation method includes detecting an operational state within a time period during which an information processing terminal displays a content; calculating, for each of parameter values related to the operational state, a level of interest in the content for each of unit time periods based on the detected operational state and the parameter values; calculating, for each of the parameter values, amounts of change in the level of interest during a plurality of time intervals arranged in chronological order based on the level of interest calculated for each of the unit time periods; and evaluating the level of interest in the content by comparing the calculated amounts of change in the interest levels for each of the parameter values with the calculated amounts of change in the level of interest for another parameter value.

Description

CROSS-REFERENCE TO RELATED APPLICATION
• This application is based upon and claims the benefit of priority of the prior Japanese Patent Application No. 2017-97641, filed on May 16, 2017, the entire contents of which are incorporated herein by reference.
FIELD
• The embodiments discussed herein are related to an interest level evaluation method, an interest level evaluating device and a recording medium.
BACKGROUND
• There is a method for analyzing “where” in a web page that a large number of unspecified users who browse the web page are interested. “Where” in a web page that users access does the users have interest and “where” in the web page is the cause for the users to leave the web page are valuable as customer information.
• The inbound marketing method for increasing potential purchase intention of persons who visit web pages and causing actual purchase behaviors is known. In the inbound marketing method, an access analysis technique for measuring interest levels indicating the degrees of interest in contents from users' behaviors on web pages is used. By taking advantaging of the interest levels, persons who have created the web pages may confirm whether the web pages attract interest as expected and may improve the web pages so that the web pages attract more interest. By taking advantaging of the interest levels, it is possible to promote users who have visited the web pages to receive a service of estimating targets in which the users are interested and recommending documents. Recently, the number of times of access to web pages from smartphones is increasing. If customer information is acquired from users who use smartphones, the number of contents valuable for the users may be increased.
• In the aforementioned technique, browsing time periods are basically measured, and interest levels are measured based on the lengths of the measured browsing time periods. For example, a technique for calculating a browsing time periods (normalized browsing time periods) per unit amount of information of a document within a content and estimating an interest score as an interest level based on the time period for browsing the document within a time interval is known.
• Especially, a technique for detecting a certain region within a content, which is fixedly laid out and is electronic newspaper or an electronic magazine, and detecting, based on an enlargement operation, a region that is within the content and in which a user is interested is known. As related art, Japanese Laid-open Patent Publication No. 2012-238114, Japanese Laid-open Patent Publication No. 2010-237942, and the like have been disclosed, for example.
• In the aforementioned technique for estimating interest levels, levels of interest in segment units of prescribed contents during time intervals may be detected. The interest levels detected in the aforementioned technique depend on the contents, and a database to be used to detect the interest levels or the like is to be created based on details of the contents. Even a method for detecting, based on an enlargement operation, a region that is included in a content and in which a user is interested has a problem with layout dependency. Especially, in a smartphone, the layout may be automatically changed due to the limitation of the horizontal width of the smartphone, and the limitation of the horizontal width varies depending on the resolution of the device. Thus, an enlargement operation varies depending on the type of the device, and it is difficult to uniformly detect interest levels.
• As described above, the conventional method for measuring interest levels depend on contents, and if the contents are variously recreated, the resolutions of portions that attract interest are increased, and levels of interest in the contents are evaluated, it takes so much time to evaluate the interest levels. Under such circumstances, it is preferable to reduce a load to be applied to a process of evaluating levels of users' interest in contents.
SUMMARY
• According to an aspect of the invention, an interest level evaluation method executed by a processor of a computer, the interest level evaluation method includes detecting an operational state within a time period during which an information processing terminal displays a content; calculating, for each of a plurality of parameter values related to the operational state, a level of interest in the content for each of a plurality of unit time periods based on the detected operational state and the plurality of parameter values; calculating, for each of the plurality of parameter values, amounts of change in the level of interest during a plurality of time intervals arranged in chronological order based on the level of interest calculated for each of the plurality of unit time periods; and evaluating the level of interest in the content by comparing the calculated amounts of change in the interest levels for each of the plurality of parameter values with the calculated amounts of change in the level of interest for another parameter value.
• The object and advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the claims.
• It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF DRAWINGS
• FIG. 1 is a block diagram schematically illustrating a configuration of an interest level evaluation system according to a first embodiment;
• FIG. 2 is a diagram illustrating an example of a method for setting a weight coefficient for a scrolling velocity;
• FIG. 3 is a diagram illustrating an example of the case where an integrated value of the amount of change in a calculated interest level is expressed as a graph;
• FIG. 4 is a diagram illustrating an example of graphs indicating amounts of change in interest levels as integrated values in the case where the amounts of change in the interested levels are calculated for parameters indicating scrolling velocities;
• FIG. 5 is a diagram illustrating an example of time intervals over a threshold and identified parameter values stored in a user information storage section according to the first embodiment;
• FIG. 6 is a block diagram schematically illustrating a configuration of a computer that functions as an information processing terminal according to the first embodiment;
• FIG. 7 is a block diagram schematically illustrating a configuration of a computer that functions as a content server according to the first embodiment;
• FIG. 8 is a flowchart illustrating an example of an interest level evaluation process according to the first embodiment;
• FIG. 9 is a block diagram schematically illustrating a configuration of an interest level evaluation system according to a second embodiment;
• FIG. 10 is a diagram illustrating an example of a method for setting weight coefficients for a terminal momentum;
• FIG. 11 is a diagram illustrating an example of time intervals over a threshold and identified parameter values stored in a user information storage section according to the second embodiment;
• FIG. 12 is a block diagram schematically illustrating a configuration of a computer that functions as an information processing terminal according to the second embodiment;
• FIG. 13 is a block diagram schematically illustrating a configuration of a computer that functions as a user information managing server according to the second embodiment;
• FIG. 14 is a flowchart illustrating an example of an interest level evaluation process according to the second embodiment;
• FIG. 15 is a block diagram schematically illustrating a configuration of an interest level evaluation system according to a third embodiment;
• FIG. 16 is a diagram illustrating an example of a method for setting weight coefficients for a screen enlargement rate;
• FIG. 17 is a block diagram schematically illustrating a configuration of a computer that functions as an information processing terminal according to the third embodiment;
• FIG. 18 is a block diagram schematically illustrating a configuration of a computer that functions as a user information managing server according to the third embodiment; and
• FIG. 19 is a flowchart illustrating an example of an interest level evaluation process according to the third embodiment.
DESCRIPTION OF EMBODIMENTS
• Hereinafter, an example of embodiments is described in detail with reference to the accompanying drawings.
• Traditionally, to evaluate an interest level indicating the degree of user's interest in a content, an interest level is evaluated for each segment of the content. Thus, if the content is recreated, it takes time and effort and is costly. In the conventional technique, the evaluation of interest levels depends on contents.
• To solve the aforementioned problems, the following two principles are introduced and interest levels are evaluated using a method that does not depend on contents. The following two principles are based on the assumption that a state in which a user becomes “interested in a content” is a psychographic change, and the psychographic change appears as a variation in a reading manner.
• As the first principle, an interest level obtained by quantifying an operational state of an information processing terminal of a user is introduced as a level that changes as time passes. By using time-series data of operational states of the information processing terminal instead of using segments of contents, interest levels may be evaluated without dependency on contents. For example, a portion that is included in a content and attracts user's interest and a portion that is included in the content and does not attract user's interest are detected based on operational states such as input operations (mouse scrolling, tapping, and the like) of continuously reading a web page or the momentum of the terminal. The portion that attracts user's interest may be used for digital marketing as customer information fed back from the user for the content of the web page.
• As the second principle, characteristic patterns that appear in time-series data of interest levels are paid attention. This is due to the fact that when a user becomes interested in a content, a characteristic pattern that is not monotonous appears in time-series data after a monotonous pattern by a monotonous page-scrolling operation. The characteristic pattern is obtained from the collection and analysis of actual data. Thus, by detecting the characteristic pattern in the time-series data, it is possible to detect a content portion attracting user's interest and identify characteristics of the user.
• The embodiments are based on the aforementioned two principles and described below.
First Embodiment
• An interestlevel evaluation system10 illustrated inFIG. 1 includes acontent server12 and aninformation processing terminal16. Thecontent server12 and theinformation processing terminal16 are connected to each other via anetwork14 such as the Internet, for example.
• Thecontent server12 transmits a content to theinformation processing terminal16 in response to a content request signal from theinformation processing terminal16.
• Theinformation processing terminal16 includes acommunication section18, acontroller20, adisplay section22, anoperation detector24, ascroll detector26, aparameter storage section28, an interestlevel quantifying section30, a temporal change comparing and detecting section32, and a userinformation storage section33. Theinformation processing terminal16 is an example of an interest level evaluating device. The interestlevel quantifying section30 and the temporal change comparing and detecting section32 are an example of an interest level evaluator.
• Thecommunication section18 transmits and receives information to and from thecontent server12. For example, thecommunication section18 receives the content transmitted by thecontent server12. Thecommunication section18 transmits, to thecontent server12, the content request signal output by thecontroller20 described later. Thecommunication section18 may periodically transmit information stored in the userinformation storage section33 to an external server for user management.
• Thecontroller20 controls the display section22 (described later) to cause thedisplay section22 to display the content received by thecommunication section18. Thecontroller20 controls thedisplay section22 based on user operation information (described later) detected by theoperation detector24 and a scrolling amount (described later) detected by thescroll detector26 so that a content desired by a user is displayed.
• Thedisplay section22 is achieved by a liquid crystal display (LCD), an organic electroluminescence display (OELD), or the like, for example. Thedisplay section22 displays the content based on control by thecontroller20. It is sufficient if the content provided by thecontent server12 is able to be displayed by thedisplay section22. The content provided by thecontent server12 may include a content including a text such as a document or a content including an image.
• Theoperation detector24 receives an input operation by the user from a touch panel superimposed on thedisplay section22 and detects the input operation by the user within a time period during which theinformation processing terminal16 displays the content. Specifically, theoperation detector24 detects the type of the input operation by the user, such as a tap operation, a flip operation, a swipe operation, a pinch operation, or the like. Theoperation detector24 detects the time when the input operation is performed and coordinates of a position at which a user's finger contacts the touch panel. Theoperation detector24 detects a scroll operation of scrolling a screen from the type of the input operation. Then, theoperation detector24 measures an operation time period within each of unit time periods. Each operation time period includes a scroll operation time period during which a scroll operation is performed. Theoperation detector24 measures a no-operation time period that is included in each of the unit time periods and during which any operation is not performed. Thus, theoperation detector24 detects operation information including the type of an input operation, the time when the input operation is performed, a contact position, an operation time period within each of the unit time periods, and a no-operation time period, which are an example of operational states of theinformation processing terminal16.
• Thescroll detector26 calculates a scrolling velocity that is an example of an operational state of theinformation processing terminal16 during a time period for a scroll operation detected by theoperation detector24. The scrolling velocity indicates the velocity of scrolling on the screen on which the content is displayed. Specifically, thescroll detector26 detects the amount (pixels) of the scrolling input by the user on the screen from the touch panel superimposed on thedisplay section22. Then, thescroll detector26 divides the amount of the scrolling by the time period (seconds) for the scroll operation detected by theoperation detector24 and calculates the scrolling velocity (pixels per second (pixels/s)) per unit time period.
• In theparameter storage section28, multiple parameter values to be used to determine a threshold to be used for the interest level quantifying section30 (described later) to calculate interest levels are stored. If the stored parameter values indicate scrolling velocities, the multiple parameter values are 70 (pixels/s), 150 (pixels/s), 500 (pixels/s), and the like.
• The interestlevel quantifying section30 calculates interest levels during each of the unit time periods based on a time period for a scroll operation detected by theoperation detector24, a no-operation time period, and a scrolling velocity calculated by thescroll detector26. The interestlevel quantifying section30 calculates the interest levels for each of the parameter values stored in theparameter storage section28 and indicating scrolling velocities. The following exemplifies the case where 70 (pixels/s) and 500 (pixels/s) are used as parameter values, and interest levels are calculated using the parameter values.
• For example, the interestlevel quantifying section30 determines a weight coefficient for a scrolling velocity detected by thescroll detector26, based on the threshold determined based on a parameter value stored in theparameter storage section28 and indicating a scrolling velocity. A method for determining the scrolling velocity weight coefficient is described later. Then, the interestlevel quantifying section30 calculates a level I(t) of user's interest in the content according to the following Equation (1) based on a scroll operation time period, a no-operation time period, and the scrolling velocity weight coefficient. The scrolling velocity weight coefficient is an example of a change in a scrolling velocity from a calculated velocity of a first scroll operation to a calculated velocity of a second scroll operation.
• $\begin{array}{cc}I\left(t\right)=\sum \frac{A\mathrm{time}\mathrm{period}\mathrm{for}a\mathrm{scroll}\mathrm{operation}}{\mathrm{The}\mathrm{scrolling}\mathrm{velocity}\mathrm{weight}\mathrm{coefficient}w_{\mathrm{scr}}}+\sum A\mathrm{no}\text{-}\mathrm{operation}\mathrm{time}\mathrm{period}& \mathrm{Equation}\left(1\right)\end{array}$
• In the first embodiment, the level I(t) of interest in the content is calculated for each unit time period (of 1 second as an example). In the aforementioned Equation (1), the “scroll operation time period” indicates a time period for a scroll operation within a unit time period, and the “no-operation time period” indicates a no-operation time period within the unit time period. The interest level I(t) is calculated in the form of an integrated value during each of the unit time periods from the time when the display of the content is started to time t.
• FIG. 2 illustrates an example of a method for setting the scrolling velocity weight coefficient w_scrfor each of the unit time periods. The scrolling velocity weight coefficient w_scris included in the aforementioned Equation (1). As illustrated inFIG. 2, the velocity weight coefficient w_scris set so that the scrolling velocity weight coefficient w_scris fixed value if a scrolling velocity v_ais in a range of 0 to a threshold v_th. The scrolling velocity weight coefficient w_scrmay be set in advance so that if the scrolling velocity v_ais equal to or larger than the threshold v_th, the scrolling velocity weight coefficient w_scrincreases with an increase in the scrolling velocity v_a. For example, as illustrated inFIG. 2, the scrolling velocity weight coefficient w_scris set so that when the scrolling velocity v_aincreases from v_thto 2v_th, the weight coefficient w_scrfor the scrolling velocity v_aincreases from A to B. Specifically, in the aforementioned Equation (1), as the scrolling velocity is higher, the scrolling velocity weight coefficient w_scris larger. As a result, the interest level during the scroll operation is evaluated to be low. Similarly, as a time period during the stop of the scroll operation is shorter, the interest level is evaluated to be lower. Specifically, if the reading of the content is skipped, the interest level is evaluated to be low in Equation (1). On the other hand, if the scroll operation is performed at a low speed so that eyeballs are able to be traced or the user reads the content while stopping the scroll operation, the interest level is evaluated to be high in Equation (1). For example, if a scrolling velocity of 70 (pixels/s) is used as a parameter value, the interest level is calculated in a state in which the scrolling velocity of 70 (pixels/s) is used as the threshold v_thand in which if the scrolling velocity is equal to or higher than 70 (pixels/s), the scrolling velocity weight coefficient is increased with an increase in the scrolling velocity. If a scrolling velocity of 500 (pixels/s) is used as the parameter value, the interest level is calculated in a state in which the scrolling velocity of 500 (pixels/s) is used as the threshold v_thand in which if the scrolling velocity is equal to or higher than 500 (pixels/s), the scrolling velocity weight coefficient is increased with an increase in the scrolling velocity.
• The interestlevel quantifying section30 calculates, as the amount of change in an interest level during each of time intervals arranged in chronological order, the difference between an interest level calculated during a unit time period including a previous time point and an interest level calculated during a unit time period including the current time point.FIG. 3 illustrates an example of the case where a change in an interest level over time is calculated in the form of an integrated value and expressed as a graph. In the graph illustrated inFIG. 3, an inclination during a time interval from a certain unit time period to the next unit time period indicates the amount of change in the interest level. In the graph illustrated inFIG. 3, it may be said that a time interval during which the interest level rapidly increases exists and a characteristic pattern of the amount of change in the interest level appears during the time interval. A scrolling velocity detected during the unit time period including the previous time point is an example of the velocity of the first scroll operation. A scrolling velocity detected during the unit time period including the current time point is an example of the velocity of the second scroll operation.
• The temporal change comparing and detecting section32 compares change amounts calculated for each of all the parameter values during each of the time intervals arranged in chronological order with change amounts calculated for the other parameter values during each of the time intervals. The temporal change comparing and detecting section32 detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than a predetermined threshold. The temporal change comparing and detecting section32 compares change amounts during a time interval preceding the detected same time interval during which the change amounts are equal to or larger than the predetermined threshold. Then, the temporal change comparing and detecting section32 identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. Then, the temporal change comparing and detecting section32 causes the detected same time interval (time interval over the threshold) and the identified parameter value to be stored in the userinformation storage section33. In this case, the time interval over the threshold and the identified parameter value are associated with a user ID, a content ID, and access time and stored in the userinformation storage section33. The access time may be a time point when an operation performed on the content has been initially detected. The temporal change comparing and detecting section32 causes the position, associated with time-series data, of the content on the display screen to be stored in the userinformation storage section33. The temporal change comparing and detecting section32 may compare not only the change amounts during the preceding time interval but also change amounts during a time interval succeeding the detected same time interval and identify parameter values. The temporal change comparing and detecting section32 detects the same time interval during which the change amounts calculated for the two or more parameter values are equal to or larger than the predetermined threshold. The temporal change comparing and detecting section32, however, may detect the same time interval during which change amounts calculated for three or more parameter values are equal to or larger than the predetermined threshold, depending on the number of the parameter values.
• FIG. 4 illustrates an example of a graph indicating changes in interest levels over time in the case where the amounts of change in the interest levels over time are calculated for parameter values indicating the scrolling velocities of 70 (pixels/s) and 500 (pixels/s). As illustrated inFIG. 4, the temporal change comparing and detecting section32 compares the change amounts for the pair of parameter values of 70 (pixels/s) and 500 (pixels/s) during the time intervals. The temporal change comparing and detecting section32 detects, as a result of the comparison, the same time interval during which change amounts per unit time period are large or gradients of the change amounts are steep for the parameter values or the time interval (time interval during which the interest levels are high) during which the interest levels rapidly increase in the graph or the amounts of change in the interest levels are large. Then, the temporal change comparing and detecting section32 also compares change amounts during a time interval preceding the time interval during which the gradients are steep. During the time interval preceding the time interval during which the gradients are steep, gradients are gentler than those during the time interval during which the change amounts are equal to or larger than the threshold. The temporal change comparing and detecting section32 identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. During a time interval during which a steep gradient that indicates a high interest level appears, a characteristic change in a change amount due to a change in a parameter value does not exist. However, during a time interval during which the interest level is low and that precedes a time period during which a steep gradient appears, changes in the interest levels are different for the parameter values. In the example illustrated inFIG. 4, during the time interval preceding the time interval during which the gradients are steep, the amount of change for the parameter value of 500 (pixels/s) is larger than the amount of change for the parameter value of 70 (pixels/s). It is apparent that a characteristic pattern of change amounts for the parameter value of 500 (pixels/s) appears, while the characteristic pattern does not appear for the parameter value of 70 (pixels/s). Specifically, parameter values specific to the user are clarified by comparing change amounts calculated for multiple parameter values. Based on the identified parameter value, it is possible to capture psychographic characteristics specific to the user or characteristics (whether the scrolling velocity is high or low and the like) of operations by the user.
• The first embodiment exemplifies the case where the number of parameter values is 2 for convenience of description. Interest levels and amounts of change in the interest levels may be calculated for three or more parameter values, and the calculated amounts of change in the interest levels may be compared.
• In the userinformation storage section33, time intervals over the threshold and identified parameter values are associated with user IDs, content IDs, and access time and stored as user-specific information. For example, a table5A illustrated inFIG. 5 indicates that a user who has a user ID “AA1” browses a content having a content ID “WWW1” at time “yy:mm:dd1:tt1” and that time intervals over the threshold detected by the temporal change comparing and detecting section32 upon the browsing are “CC1 and CC2” and a parameter value identified by the temporal change comparing and detecting section32 upon the browsing is “D2”. In the userinformation storage section33, the positions, associated with the time-series data, of contents on the display screen are stored.
• Theinformation processing terminal16 may be achieved by acomputer50 illustrated inFIG. 6, for example. Thecomputer50 includes aCPU51, amemory52 as a temporal storage region, and anonvolatile storage section53. Thecomputer50 further includes an input andoutput device54 including thedisplay section22 and the touch panel superimposed on thedisplay section22, and a reading and writing (R/W)section55 that controls reading and writing of data from and to arecording medium59. Thecomputer50 further includes a network interface (I/F)56 that is connected to the network such as the Internet. TheCPU51, thememory52, thestorage section53, the input andoutput device54, the R/W section55, and the network I/F56 are connected to each other via abus57.
• Thestorage section53 may be achieved by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, or the like. In thestorage section53 serving as a storage medium, an interestlevel evaluation program60 that causes thecomputer50 to function as theinformation processing terminal16 is stored. The interestlevel evaluation program60 includes acommunication process62, acontrol process63, anoperation detection process65, ascroll detection process66, acalculation process67, and acomparison process68. Thestorage section53 includes aparameter storage region69 in which information constituting theparameter storage section28 is stored. Thestorage section53 includes a userinformation storage region71 in which information constituting the userinformation storage section33 is stored.
• TheCPU51 reads the interestlevel evaluation program60 from thestorage section53, loads the read interestlevel evaluation program60 into thememory52, and sequentially executes the processes included in the interestlevel evaluation program60. TheCPU51 executes thecommunication process62, thereby operating as thecommunication section18 illustrated inFIG. 1. TheCPU51 executes thecontrol process63, thereby operating as thecontroller20 illustrated inFIG. 1. TheCPU51 executes theoperation detection process65, thereby operating as theoperation detector24 illustrated inFIG. 1. TheCPU51 executes thescroll detection process66, thereby operating as thescroll detector26 illustrated inFIG. 1. TheCPU51 executes thecalculation process67, thereby operating as the interestlevel quantifying section30 illustrated inFIG. 1. TheCPU51 executes thecomparison process68, thereby operating as the temporal change comparing and detecting section32 illustrated inFIG. 1. TheCPU51 reads the information from theparameter storage region69 and loads theparameter storage section28 into thememory52. Thus, theCPU51 executes the interestlevel evaluation program60, thereby functioning as theinformation processing terminal16. TheCPU51 that executes the program is hardware.
• The functions that are achieved by the interestlevel evaluation program60 may be achieved by a semiconductor integrated circuit, or more specifically, by an application specific integrated circuit (ASIC) or the like, for example.
• Thecontent server12 may be achieved by acomputer80 illustrated inFIG. 7, for example. Thecomputer80 includes aCPU81, amemory82 as a temporal storage region, and anonvolatile storage section83. Thecomputer80 further includes an input andoutput device84 including a display device, an input device, and the like, and an R/W section85 that controls reading and writing of data from and to arecording medium89. Thecomputer80 further includes a network I/F86 that is connected to the network such as the Internet. TheCPU81, thememory82, thestorage section83, the input andoutput device84, the R/W section85, and the network I/F86 are connected to each other via abus87.
• Thestorage section83 may be achieved by an HDD, an SSD, a flash memory, or the like. In thestorage section83 serving as a storage medium, a content provision program90 that causes thecomputer80 to function as thecontent server12 is stored. In a content storage region98, contents able to be provided to theinformation processing terminal16 are stored in advance.
• The function that is achieved by the content provision program90 may be achieved by a semiconductor integrated circuit, or more specifically, by an ASIC or the like, for example.
• Next, effects of the interestlevel evaluation system10 according to the first embodiment are described. In the interestlevel evaluation system10, theinformation processing terminal16 receives a content from thecontent server12. Then, the received content is displayed by thedisplay section22 of theinformation processing terminal16. When theoperation detector24 receives an input operation by a user, an interest level evaluation process illustrated inFIG. 8 is executed by theinformation processing terminal16. Processes are described below.
• In S100, theoperation detector24 detects a time period for a scroll operation by the user within a unit time period. Thescroll detector26 detects the amount of the scrolling input by the user on the screen from the touch panel superimposed on thedisplay section22.
• In S101, thescroll detector26 divides the detected amount of the scrolling by the scroll operation time period detected in the aforementioned S100 within the unit time period and calculates the scrolling velocity (pixels/s).
• In S102, the interestlevel quantifying section30 determines whether or not the display of the content has been terminated. If the display has been terminated, the process proceeds to S103. If the display has not been terminated, the process returns to S100 and is repeated.
• In S103, the interestlevel quantifying section30 selects a parameter value stored in theparameter storage section28 and indicating a scrolling velocity.
• In S104, the interestlevel quantifying section30 calculates, for the parameter value selected in S103, an interest level during each of unit time periods. The interest level during each of the unit time periods is calculated according to the aforementioned Equation (1) based on the scroll operation time period detected in S100, a no-operation time period, and the scrolling velocity calculated in S101.
• In S105, the interestlevel quantifying section30 calculates, for the parameter value selected in S103, amounts of change in the interest levels during the unit time periods. Each of the change amounts is calculated as the difference between an interest level calculated during a unit time period that is included in the time intervals arranged in chronological order and includes a previous time point and an interest level calculated during a unit time period that is included in the time intervals arranged in chronological order and includes the current time point.
• In S106, it is determined whether interest levels and amounts of change in the interest levels have been calculated for the all parameter values stored in theparameter storage section28. If the interest levels and the amounts of change in the interest levels have been calculated for all the parameter values as a result of the determination, the process proceeds to S107. On the other hand, if interest levels and the amounts of change in the interest levels have not been calculated for one or more of all the parameter values, the process returns to S103, the next parameter value is selected, and the process is repeated.
• In S107, the temporal change comparing and detecting section32 compares the change amounts calculated for each of the parameter values during the time intervals arranged in chronological order with the change amounts calculated for the other parameter values during the time intervals arranged in chronological order. Then, the temporal change comparing and detecting section32 detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than the predetermined threshold.
• In S108, the temporal change comparing and detecting section32 compares change amounts during a time interval preceding the detected same time interval during which the change amounts are equal to or larger than the predetermined threshold, and identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated.
• In S109, the temporal change comparing and detecting section32 causes the same time interval detected in S107 and the parameter value identified in S108 to be stored in the userinformation storage section33. In this case, the information is stored while being associated with information (user ID, content ID, and access time) specific to the user who has browsed the content.
• As described above, upon receiving an input operation within a time period during which the information processing terminal displays a content, the interest level evaluation system according to the first embodiment detects operational states including the velocity of a scroll operation. Then, the information processing terminal calculates interest levels for the parameter values based on the detected operational states and calculates amounts of change in the interest levels during the time intervals arranged in chronological order. Then, the information processing terminal compares the amounts of change in the interest levels and detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than the predetermined threshold. Then, the information processing terminal compares change amounts during a time interval preceding the detected same time interval and identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. Consequently, evaluation precision of the levels of users' interest in contents may be improved without depending on the contents.
Second Embodiment
• Next, a second embodiment is described. Sections that are the same as those described in the first embodiment are indicated by the same reference symbols as those described in the first embodiment, and a description thereof is omitted.
• The second embodiment is different from the first embodiment in that the momentum of an information processing terminal is used for interest level evaluation in the second embodiment. The second embodiment is different from the first embodiment in that the interest level evaluation is executed by a user information managing server in the second embodiment. The momentum of the terminal is an example of a “motion”.
• An interestlevel evaluation system210 illustrated inFIG. 9 according to the second embodiment includes thecontent server12, aninformation processing terminal216, and a userinformation managing server213. Thecontent server12, theinformation processing terminal216, and the userinformation managing server213 are connected to each other via thenetwork14 such as the Internet.
• Theinformation processing terminal216 includes thecommunication section18, thecontroller20, thedisplay section22, theoperation detector24, thescroll detector26, and aterminal momentum detector227.
• Theterminal momentum detector227 detects the momentum of theinformation processing terminal216 for each of the unit time periods. The terminal momentum is an example of an operational state of theinformation processing terminal216. The second embodiment describes the case where theterminal momentum detector227 is achieved by a 9-axis sensor. The 9-axis sensor is composed of three types of sensors, a triaxial gyro sensor, a triaxial acceleration sensor, and a triaxial geomagnetic sensor. Theterminal momentum detector227 may be achieved by one or more of the three types of sensors. In this case, the unit time periods are predetermined detection cycles T (of 1 second as an example).
• Thecommunication section18 according to the second embodiment transmits, to the userinformation managing server213 based on a control process by thecontroller20, operation information including the type of an input operation detected by theoperation detector24, the time when the input operation has been performed, a contact position, a time period for the operation, and a no-operation time period. In the second embodiment, the operation information also includes a scrolling velocity detected by thescroll detector26. Thecommunication section18 transmits, to the userinformation managing server213, the terminal momentum detected by theterminal momentum detector227. The operation information to be transmitted and the terminal momentum to be transmitted are associated with a user ID, a content ID of a content being displayed, and detection time for each of the unit time periods. Thecommunication section18 also transmits the position, associated with time-series data, of the content on the display screen.
• The userinformation managing server213 includes acommunication section218, an operationalstate storage section219, aserver controller220, aparameter storage section228, an interestlevel quantifying section230, a temporal change comparing and detectingsection232, and a userinformation storage section233.
• Thecommunication section218 transmits and receives information to and from theinformation processing terminal216. For example, thecommunication section218 receives, from theinformation processing terminal216, the operation information and terminal momentum associated with the user ID, the content ID, and the detection time. Thecommunication section218 receives the position, associated with the time-series data, of the content on the display screen.
• Theserver controller220 acquires the operation information and terminal momentum received by thecommunication section218 and causes the acquired operation information and the acquired terminal momentum to be stored in the operationalstate storage section219. Theserver controller220 notifies the interestlevel quantifying section230 that the newly received operation information and the newly received terminal momentum have been stored in the operationalstate storage section219. Theserver controller220 causes the position, associated with the time-series data, of the content on the display screen to be stored in the userinformation storage section233.
• In the operationalstate storage section219, terminal momentums and operation information that are associated with user IDs, content IDs, and detection time are stored.
• In theparameter storage section228, multiple parameter values to be used to determine thresholds to be used for the interest level quantifying section230 (described later) to quantify a change in the terminal momentum are stored. If the stored parameter values indicate scrolling velocities, the multiple parameter values are, for example, 70 (pixels/s), 150 (pixels/s), 500 (pixels/s), and the like. If the multiple parameter values indicate terminal momentums, the multiple parameter values are, for example, 0.05 (rad/s²), 0.01 (rad/s²), 0.0025 (rad/s²), and the like. Each of the values of the terminal momentums stored as the parameter values is obtained by acquiring and measuring a rotational momentum of the information processing terminal five times per second and calculating the sum of squares of the rotational momentums.
• The interestlevel quantifying section230 calculates interest levels for each of the unit time periods based on the operation time periods, no-operation time periods, and terminal momentums included in the operation information stored in the operationalstate storage section219. The interestlevel quantifying section230 calculates the aforementioned interest levels for the multiple parameter values stored in theparameter storage section228 and indicating the terminal momentums. The interestlevel quantifying section230 may calculate interest levels for the multiple parameter values indicating the scrolling velocities in the same manner as the aforementioned first embodiment and calculate interest levels for the multiple parameter values indicating the terminal momentums.
• For example, the interestlevel quantifying section230 determines weight coefficients for the terminal momentum based on the thresholds determined based on the parameter values stored in theparameter storage section228 and indicating the terminal momentums. Then, the interestlevel quantifying section230 calculates a level I(t) of user's interest in the content according to the following Equation (2) based on an operation time period, a no-operation time period, and the weight coefficients for the terminal momentum. The weight coefficients for the terminal momentum are an example of a change in the terminal momentum.
• $\begin{array}{cc}I\left(t\right)=\sum \frac{\mathrm{An}\mathrm{operation}\mathrm{time}\mathrm{period}}{\mathrm{The}\mathrm{terminal}\mathrm{momentum}\mathrm{weight}\mathrm{coefficient}w_d}+\sum \frac{A\mathrm{no}\text{-}\mathrm{operation}\mathrm{time}\mathrm{period}}{\mathrm{The}\mathrm{terminal}\mathrm{momentum}\mathrm{weight}\mathrm{coefficient}w_{\mathrm{nop}}}& \mathrm{Equation}\left(2\right)\end{array}$
• In the second embodiment, the level I(t) of interest in the content is calculated for each of the unit time periods (of 1 second as an example). In the aforementioned Equation (2), the “operation time period” is a time period for an operation within a unit time period, and a “no-operation time period” is a time period that is included in the unit time period and during which any operation is not performed. The weight coefficients for the terminal momentum are set as a terminal momentum weight coefficient w_dfor the operation time period and a terminal momentum weight coefficient w_nopfor the no-operation time period. For example, as illustrated inFIG. 10, the terminal momentum weight coefficient w_dfor the operation time period is calculated from the kinetic power of the information processing terminal during the operation time period included in the unit time period and a threshold PA_avefor the momentum during the operation time period. As illustrated inFIG. 10, the terminal momentum weight coefficient w_nopfor the no-operation time period is calculated from the kinetic power of the information processing terminal during the no-operation time period included in the unit time period and a threshold PB_avefor the momentum during the no-operation time period. In the following description, the terminal momentum weight coefficients w_dand w_nopare referred to as weight coefficients w_dand w_nop.
• If a parameter value is set to 0.05 (rad/s²), the weight coefficient w_dand the weight coefficient w_nopare set so that when the parameter value increases by 0.05 (rad/s²), the weight coefficients increase by 1 for the terminal momentum exceeding the threshold. If the parameter value is set to 0.01 (rad/s²), the weight coefficient w_dand the weight coefficient w_nopare set so that when the parameter value increases by 0.01 (rad/s²), the weight coefficients increase by 1 for the terminal momentum exceeding the threshold. Specifically, amounts of increase in the weight coefficients are set based on the parameter value.
• The threshold PA_avefor the kinetic power of the information processing terminal during the operation time period and the threshold PB_avefor the kinetic power of the information processing terminal during the no-operation time period are determined based on a parameter value stored in theparameter storage section228 and indicating a terminal momentum. For example, the parameter value may be assigned to the threshold PB_ave, and the threshold PA_avemay be set to a value obtained by adding a predetermined value to the threshold PB_ave.
• According to the weight coefficient w_dindicated in the aforementioned Equation (2) and illustrated inFIG. 10, if the kinetic power of the information processing terminal during the operation time period is equal to or smaller than the threshold PA_ave, the weight coefficient w_dis equal to 1.0, and the operation time period is added to an evaluated value I(t) of the interest level. Thus, if the kinetic power of the information processing terminal during the operation time period is small and it is estimated that the level of user's interest in the content is high, the evaluated value I(t) of the interest level is high.
• If the kinetic power of the information processing terminal during the operation time period exceeds the threshold PA_ave, the weight coefficient w_dincreases with an increase in the kinetic power, and an effect of the operation time period on the evaluated value I(t) of the interest level is small. Thus, if the kinetic power of the information processing terminal during the operation time period is large and it is estimated that the user does not concentrate on browsing the content, the evaluated value I(t) of the interest level is low.
• According to the weight coefficient w_nopindicated in the aforementioned Equation (2) and illustrated inFIG. 10, if the kinetic power of the information processing terminal during the no-operation time period is equal to or smaller than the threshold PB_ave, the weight coefficient w_nopis equal to 1.0 and the no-operation time period is added to the evaluated value I(t) of the interest level. Thus, if the kinetic power of the information processing terminal during the no-operation time period is small and it is estimated that the level of user's interest in the content is high, the evaluated value I(t) of the interest level is high.
• If the kinetic power of the information processing terminal during the no-operation time period exceeds the threshold PB_ave, the value of the weight coefficient w_nopincreases with an increase in the kinetic power, and an effect of the no-operation time period on the evaluated value I(t) of the interest level is small. Thus, if the kinetic power of the information processing terminal during the no-operation time period is large and it is estimated that the user does not concentrate on browsing the content, the evaluated value I(t) of the interest level is low.
• Similarly to the first embodiment, the interestlevel quantifying section230 calculates, for each of the parameter values indicating the terminal momentums, the difference between an interest level calculated during a unit time period including a previous time point and an interest level calculated during a unit time period including the current time point, as the amount of change in the interest level during each of the time intervals arranged in chronological order.
• The temporal change comparing and detectingsection232 compares the change amounts calculated for each of the parameter values during the time intervals arranged in chronological order with the change amounts calculated for the other parameter values during the time intervals arranged in chronological order. Then, the temporal change comparing and detectingsection232 detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than the predetermined threshold. The temporal change comparing and detectingsection232 compares change amounts during a time interval preceding the detected same time interval during which the change amounts are equal to or larger than the predetermined threshold. Then, the temporal change comparing and detectingsection232 identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. Then, the temporal change comparing and detectingsection232 associates the detected same time interval (time interval over the threshold) during which the change amounts are equal to or larger than the predetermined threshold with a user ID, a content ID, and access time and causes the time interval over the threshold associated with the user ID, the content ID, and the access time to be stored in the userinformation storage section233. The temporal change comparing and detectingsection232 associates the identified parameter value (identified parameter value) for which the change amount that is larger than the change amount calculated for the other parameter value has been calculated with the user ID, the content ID, and the access time, and the temporal change comparing and detectingsection232 causes the identified parameter value associated with the user ID, the content ID, and the access time to be stored in the userinformation storage section233.
• In the userinformation storage section233, the time interval over the threshold and the identified parameter value are associated with the user ID, the content ID, and the access time and stored as user-specific information. For example, a table11A illustrated inFIG. 11 indicates that a user having a user ID “AA2” browses a content having a content ID “WWW1” at time “yy:mm:dd1:tt1” and that time intervals over the threshold detected by the temporal change comparing and detecting section32 upon the browsing are “CC1 and CC2” and a parameter value identified by the temporal change comparing and detecting section32 upon the browsing is “D2”. In the userinformation storage section233, the position, associated with time-series data, of the content on the display screen is stored.
• Theinformation processing terminal216 may be achieved by acomputer250 illustrated inFIG. 12, for example. Thecomputer250 includes aCPU51, amemory52 as a temporal storage region, and anonvolatile storage section253. Thecomputer250 further includes an input andoutput device54, an R/W section55, and a network I/F56. TheCPU51, thememory52, thestorage section253, the input andoutput device54, the R/W section55, and the network I/F56 are connected to each other via abus57.
• Thestorage section253 may be achieved by an HDD, an SSD, a flash memory, or the like. In thestorage section253 serving as a storage medium, anoperation program260 that causes thecomputer250 to function as theinformation processing terminal216 is stored. Theoperation program260 includes acommunication process62, acontrol process63, anoperation detection process65, ascroll detection process66, and amomentum detection process266.
• TheCPU51 reads theoperation program260 from thestorage section253, loads the readoperation program260 into thememory52, and sequentially executes the processes included in theoperation program260. TheCPU51 executes thecommunication process62, thereby operating as thecommunication section18 illustrated inFIG. 9. TheCPU51 executes thecontrol process63, thereby operating as thecontroller20 illustrated inFIG. 9. TheCPU51 executes theoperation detection process65, thereby operating as theoperation detector24 illustrated inFIG. 9. TheCPU51 executes thescroll detection process66, thereby operating as thescroll detector26 illustrated inFIG. 9. TheCPU51 executes themomentum detection process266, thereby operating as theterminal momentum detector227 illustrated inFIG. 9. Thus, thecomputer250 executes theoperation program260, thereby functioning as theinformation processing terminal216. TheCPU51 that executes the program is hardware.
• The functions that are achieved by theoperation program260 may be achieved by a semiconductor integrated circuit, or more specifically, by an ASIC or the like, for example.
• The userinformation managing server213 may be achieved by acomputer280 illustrated inFIG. 13, for example. Thecomputer280 includes aCPU91, amemory92 as a temporal storage region, and anonvolatile storage section283. Thecomputer280 further includes an input andoutput device94 including a display device, an input device, and the like, and an R/W section95 that controls reading and writing of data from and to arecording medium99. Thecomputer280 further includes a network I/F96 that is connected to the network such as the Internet. TheCPU91, thememory92, thestorage section283, the input andoutput device94, the R/W section95, and the network I/F96 are connected to each other via abus97.
• Thestorage section283 may be achieved by an HDD, an SSD, a flash memory, or the like. In thestorage section283 serving as a storage medium, an interestlevel evaluation program261 that causes thecomputer280 to function as the userinformation managing server213 is stored. The interestlevel evaluation program261 includes acommunication process262, acontrol process263, acalculation process267, and acomparison process268. Thestorage section283 includes aparameter storage region269 in which information constituting theparameter storage section228 is stored. Thestorage section283 includes an operationalstate storage region270 in which information constituting the operationalstate storage section219 is stored. Thestorage section283 includes a userinformation storage region271 in which information constituting the userinformation storage section233 is stored.
• TheCPU91 reads the interestlevel evaluation program261 from thestorage section283, loads the read interestlevel evaluation program261 into thememory82, and sequentially executes the processes included in the interestlevel evaluation program261. TheCPU91 executes thecommunication process262, thereby operating as thecommunication section218 illustrated inFIG. 9. TheCPU91 executes thecontrol process263, thereby operating as theserver controller220 illustrated inFIG. 9. TheCPU91 executes thecalculation process267, thereby operating as the interestlevel quantifying section230 illustrated inFIG. 9. TheCPU91 executes thecomparison process268, thereby operating as the temporal change comparing and detectingsection232 illustrated inFIG. 9. TheCPU91 reads the information from theparameter storage region269 and loads theparameter storage section228 into thememory92. TheCPU91 reads the information from the operationalstate storage region270 and loads the operationalstate storage section219 into thememory92. TheCPU91 reads the information from the userinformation storage region271 and loads the userinformation storage section233 into thememory92. Thus, thecomputer280 executes the interestlevel evaluation program261, thereby functioning as the userinformation managing server213. TheCPU91 that executes the program is hardware.
• The functions that are achieved by the interestlevel evaluation program261 may be achieved by a semiconductor integrated circuit, or more specifically, by an ASIC or the like, for example.
• Next, effects of the interestlevel evaluation system210 according to the second embodiment are described. In the interestlevel evaluation system210, theinformation processing terminal216 receives a content from thecontent server12. Then, the content is displayed by thedisplay section22 of theinformation processing terminal216. When an input operation is performed by a user, detection processes are executed by theoperation detector24, thescroll detector26, and theterminal momentum detector227 during each of the unit time periods. Then, theinformation processing terminal216 executes the detection processes until the display of the content is terminated. Then, thecommunication section18 of theinformation processing terminal216 associates, based on a control process by thecontroller20, operation information detected during each of the unit time periods and a terminal momentum detected during each of the unit time periods with a user ID, a content ID, and detection time and transmits the operation information and terminal momentum associated with the user ID, the content ID, and the detection time to the userinformation managing server213 based on a control process by thecontroller20. When thecommunication section218 of the userinformation managing server213 receives the information transmitted by theinformation processing terminal216, the userinformation managing server213 executes an interest level evaluation process illustrated inFIG. 14. Processes are described below.
• In S200, theserver controller220 causes the operation information and terminal momentum detected during each of the unit time periods and received by thecommunication section218 and associated with the user ID, the content ID, and the detection time to be stored in the operationalstate storage section219. Theserver controller220 notifies the interestlevel quantifying section230 that the information has been stored in the operationalstate storage section219.
• In S201, the interestlevel quantifying section230 selects a parameter value stored in theparameter storage section228 and indicating a terminal momentum.
• In S202, the interestlevel quantifying section230 calculates, for the parameter value selected in S201, an interest level during each of the unit time periods. The interest level during each of the unit time periods is calculated according to the aforementioned Equation (2) based on an operation time period and no-operation time period indicated in the operation information stored in the operationalstate storage section219 and the terminal momentum.
• In S105 to S109, in the same manner as the first embodiment, change amounts are calculated, the same time interval during which change amounts are equal to or larger than the predetermined threshold is detected, and a parameter value for which a change amount that is larger than a change amount calculated for another parameter value has been calculated is identified. Then, the detected time interval and the identified parameter value are stored as user-specific information in the userinformation storage section233.
• As described above, upon receiving an input operation within a time period during which the information processing terminal displays a content, the interest level evaluation system according to the second embodiment detects operational states including the terminal momentum. Then, the user information managing server calculates interest levels for the parameter values based on the detected operational states and calculates amounts of change in the interest levels during the time intervals arranged in chronological order. Then, the server information managing server compares the change amounts and detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than the predetermined threshold. Then, the user information managing server compares change amounts during a time interval preceding the detected same time interval and identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. Consequently, evaluation precision of the levels of users' interest in contents may be improved without depending on the contents.
Third Embodiment
• Next, a third embodiment is described. Sections that are the same as those described in the second embodiment are indicated by the same reference symbols as those described in the second embodiment, and a description thereof is omitted.
• The third embodiment is different from the second embodiment in that a screen enlargement rate of an information processing terminal is used for interest level evaluation in the third embodiment.
• An interestlevel evaluation system310 illustrated inFIG. 15 according to the third embodiment includes thecontent server12, aninformation processing terminal316, and a userinformation managing server313. Thecontent server12, theinformation processing terminal316, and the userinformation managing server313 are connected to each other via thenetwork14 such as the Internet, for example.
• Theinformation processing terminal316 includes thecommunication section18, thecontroller20, thedisplay section22, theoperation detector24, thescroll detector26, and a screenenlargement rate detector327.
• Thescreen enlargement detector327 detects a screen enlargement rate that is an example of an operational state of theinformation processing terminal316. The screen enlargement rate is detected based on the state of an operation of zooming in and out a screen by pinch-out and pinch-in operations among input operations detected by theoperation detector24. The screenenlargement rate detector327 detects a zoom operation time period for reduction and enlargement operations that are the pinch-in and pinch-out operations.
• Thecommunication section18 according to the third embodiment transmits, to the userinformation managing server213, operation information detected by theoperation detector24 and including the types of the input operations, time when the input operations have been performed, contact positions, and a no-operation time period, based on a control process by thecontroller20. In the third embodiment, the operation information includes a scrolling velocity detected by thescroll detector26. Thecommunication section18 transmits the screen enlargement rate detected by the screenenlargement rate detector327 and the zoom operation time period detected by the screenenlargement rate detector327 to the userinformation managing server213. The operation information, the screen enlargement rate, and the zoom operation time period that are to be transmitted are associated with a user ID, a content ID of a content being displayed, and detection time for each of the unit time periods.
• The userinformation managing server313 includes thecommunication section218, theserver controller220, an operationalstate storage section319, theparameter storage section228, an interestlevel quantifying section330, the temporal change comparing and detectingsection232, and the userinformation storage section233.
• Thecommunication section218 according to the third embodiment receives the operation information, screen enlargement rate, and zoom operation time period associated with the user ID, the content ID, and the detection time and transmitted by theinformation processing terminal316.
• Theserver controller220 according to the third embodiment acquires the operation information, screen enlargement rate, and zoom operation time period received by thecommunication section218 and causes the operation information, the screen enlargement rate, and the zoom operation time period to be stored in the operationalstate storage section319. Theserver controller220 notifies the interestlevel quantifying section330 that the newly received operation information, the newly received screen enlargement rate, and the newly received zoom operation time period have been stored in the operationalstate storage section319.
• In the operationalstate storage section319, the operation information, the screen enlargement rate, and the zoom operation time period are associated with the user ID, the content ID, and the detection time and stored.
• The interestlevel quantifying section330 calculates interest levels during each of the unit time periods based on a scroll operation time period, no-operation time period, and scrolling velocity included in the operation information stored in the operationalstate storage section319, the screen enlargement rate, and the zoom operation time period. The interestlevel quantifying section330 calculates the interest levels for each of the parameter values stored in theparameter storage section228 and indicating scrolling velocities.
• For example, the interestlevel quantifying section330 determines a weight coefficient for the scrolling velocity based on the threshold determined based on the parameter values stored in theparameter storage section228 and indicating the scrolling velocities. A method for determining the weight coefficient for the scrolling velocity is the same as or similar to that described in the first embodiment. Weight coefficients for the screen enlargement rate are determined based on the screen enlargement rate. Then, the interestlevel quantifying section330 calculates a level I(t) of user's interest in the content according to the following Equation (3) based on a scroll operation time period, a no-operation time period, the weight coefficient for the scrolling velocity, and the weight coefficients for a screen enlargement rate. The weight coefficients for the screen enlargement rate are an example of a change in the screen enlargement rate.
• $\mathrm{Equation}\left(3\right)$$I\left(t\right)=\sum A\mathrm{time}\mathrm{period}\mathrm{for}a\mathrm{zoom}\mathrm{operation}+\sum \frac{A\mathrm{time}\mathrm{period}\mathrm{for}a\mathrm{scroll}\mathrm{operation}}{\mathrm{The}\mathrm{scrolling}\mathrm{velocity}\mathrm{weight}\mathrm{coefficient}w_{\mathrm{scr}}}*\hspace{1em}\mathrm{The}\mathrm{enlargement}\mathrm{rate}\mathrm{coefficient}z_{\mathrm{scr}}+\sum A\mathrm{no}\text{-}\mathrm{operation}\mathrm{time}\mathrm{period}*\mathrm{The}\mathrm{enlargement}\mathrm{rate}\mathrm{coefficient}z_{\mathrm{nop}}$
• In the third embodiment, the level I(t) of interest in the content is calculated for each of the unit time periods (of 1 second as an example). In the aforementioned Equation (3), a method for setting the weight w_scrfor the scrolling velocity is the same as or similar to that described in the first embodiment. The weight coefficients for the screen enlargement rate are set for an enlargement rate coefficient z_scrand an enlargement rate coefficient z_nop. The enlargement rate coefficient z_sris a weight coefficient related to the screen enlargement rate and provided for an operation time period. The enlargement rate coefficient z_nopis a weight coefficient related to the screen enlargement rate and provided for a no-operation time period. For example, as illustrated inFIG. 16, it is assumed that a weight when the screen enlargement rate is 2.0 is 1.0 and that as the enlargement rate is higher, the level of user's interest in the content is higher. Based on this assumption, the enlargement rate coefficients z_scrand z_nopare set so that the enlargement rate coefficients z_scrand z_noplinearly increase with an increase in the screen enlargement rate. In an example illustrated inFIG. 16, when the screen enlargement rate is increased to 2.0, the enlargement rate coefficients z_scrand z_nopincrease to 2.0. Specifically, the enlargement rate coefficients z_scrand z_nopincrease by 1 based on a multiple of the screen enlargement rate. In the third embodiment, the rates of increase in the enlargement rate coefficients z_scrand z_nopare equal to each other. For example, based on the assumption that an interest level is high during a no-operation time period, the rate of increase in the enlargement rate coefficient z_nopmay be 1.2.
• Theinformation processing terminal316 may be achieved by acomputer350 illustrated inFIG. 17, for example. Thecomputer350 includes theCPU51, thememory52 as a temporal storage region, and anonvolatile storage section353. Thecomputer350 further includes the input andoutput device54, the R/W section55, and the network I/F56. TheCPU51, thememory52, thestorage section253, the input andoutput device54, the R/W section55, and the network I/F56 are connected to each other via thebus57.
• Thestorage section353 may be achieved by an HDD, an SSD, a flash memory, or the like. In thestorage section353 serving as a storage medium, anoperation program360 that causes thecomputer350 to function as theinformation processing terminal316 is stored. Theoperation program360 includes thecommunication process62, thecontrol process63, theoperation detection process65, thescroll detection process66, and a screen enlargementrate detection process366.
• TheCPU51 reads theoperation program360 from thestorage section353, loads the readoperation program360 into thememory52, and sequentially executes the processes included in theoperation program360. TheCPU51 executes thecommunication process62, thereby operating as thecommunication section18 illustrated inFIG. 15. TheCPU51 executes thecontrol process63, thereby operating as thecontroller20 illustrated inFIG. 15. TheCPU51 executes theoperation detection process65, thereby operating as theoperation detector24 illustrated inFIG. 15. TheCPU51 executes thescroll detection process66, thereby operating as thescroll detector26 illustrated inFIG. 15. TheCPU51 executes the screen enlargementrate detection process366, thereby operating as the screenenlargement rate detector327 illustrated inFIG. 15. Thus, thecomputer350 executes theoperation program360, thereby functioning as theinformation processing terminal316. TheCPU51 that executes the program is hardware.
• The functions that are achieved by theoperation program360 may be achieved by a semiconductor integrated circuit, or more specifically, by an ASIC or the like, for example.
• The userinformation managing server313 may be achieved by acomputer380 illustrated inFIG. 18, for example. Thecomputer380 includes theCPU91, thememory92 as a temporal storage region, and anonvolatile storage section383. Thecomputer380 further includes the input andoutput device94 and the R/W section95 that controls reading and writing of data from and to therecording medium99. Thecomputer380 further includes the network I/F96 that is connected to the network such as the Internet. TheCPU91, thememory92, thestorage section383, the input andoutput device94, the R/W95, and the network I/F96 are connected to each other via thebus97.
• Thestorage section383 may be achieved by an HDD, an SSD, a flash memory, or the like. In thestorage section383 serving as a storage medium, an interestlevel evaluation program361 that causes thecomputer380 to function as the userinformation managing server313 is stored. The interestlevel evaluation program361 includes thecommunication process262, thecontrol process263, acalculation process367, and thecomparison process268. Thestorage section383 includes theparameter storage region269 in which the information constituting theparameter storage section228 is stored. Thestorage section383 includes an operationalstate storage region370 in which information constituting the operationalstate storage section319 is stored. Thestorage section383 includes the userinformation storage region271 in which the information constituting the userinformation storage section233 is stored.
• TheCPU91 reads the interestlevel evaluation program361 from thestorage section383, loads the read interestlevel evaluation program361 into thememory92, and sequentially executes the processes included in the interestlevel evaluation program361. TheCPU91 executes thecommunication process262, thereby operating as thecommunication section218 illustrated inFIG. 15. TheCPU91 executes thecontrol process263, thereby operating as theserver controller220 illustrated inFIG. 15. TheCPU91 executes thecalculation process367, thereby operating as the interestlevel quantifying section330 illustrated inFIG. 15. TheCPU91 executes thecomparison process268, thereby operating as the temporal change comparing and detectingsection232 illustrated inFIG. 15. TheCPU91 reads the information from theparameter storage region269 and loads theparameter storage section228 into thememory92. TheCPU91 reads the information from the operationalstate storage region370 and loads the operationalstate storage section319 into thememory92. TheCPU91 reads the information from the userinformation storage region271 and loads the userinformation storage section233 into thememory92. Thus, thecomputer380 executes the interestlevel evaluation program361, thereby functioning as the userinformation managing server313. TheCPU91 that executes the program is hardware.
• The functions that are achieved by the interestlevel evaluation program361 may be achieved by a semiconductor integrated circuit, or more specifically, by an ASIC or the like, for example.
• Next, effects of the interestlevel evaluation system310 according to the third embodiment are described. In the interestlevel evaluation system310, theinformation processing terminal316 receives a content from thecontent server12. Then, the content is displayed by thedisplay section22 of theinformation processing terminal316. When an input operation is performed by a user, detection processes are executed by theoperation detector24, thescroll detector26, and the screenenlargement rate detector327 for each of the unit time periods. Then, theinformation processing terminal316 executes the detection processes until the display of the content is terminated. Then, thecommunication section18 of theinformation processing terminal316 associates, based on a control process by thecontroller20, operation information, screen enlargement rate, and zoom operation time period detected for each of the unit time periods with a user ID, a content ID, and detection time and transmits the operation information, screen enlargement rate, and zoom operation time associated with the user ID, the content ID, and the detection time to the userinformation managing server313. When thecommunication section218 of the userinformation managing server313 receives the information transmitted by theinformation processing terminal316, the userinformation managing server313 executes an interest level evaluation process illustrated inFIG. 19. Processes are described below.
• In S300, theserver controller220 causes the operation information and screen enlargement rate detected for each of the unit time periods and received by thecommunication unit218 and associated with the user ID, the content ID, and the detection time to be stored in the operationalstate storage section319. Theserver controller220 notifies the interestlevel quantifying section330 that the information has been stored in the operationalstate storage section319.
• In S301, the interestlevel quantifying section330 selects a parameter value stored in theparameter storage section228 and indicating a scrolling velocity.
• In S302, the interestlevel quantifying section330 calculates, for the selected parameter value, an interest level during each of the unit time periods. The interest level during each of the unit time periods is calculated according to the aforementioned Equation (3) based on a scroll operation time period, no-operation time period, and scrolling velocity included in the operation information stored in the operationalstate storage section319, the screen enlargement rate, and the zoom operation time period.
• In S105 to S109, in the same manner as the first embodiment, change amounts are calculated, the same time interval during which change amounts are equal to or larger than the predetermined threshold is detected, and a parameter value for which a change amount that is larger than a change amount calculated for another parameter value has been calculated is identified. Then, the detected time interval and the identified parameter value are stored as user-specific information in the userinformation storage section233.
• As described above, upon receiving an input operation within a time period during which the information processing terminal displays a content, the interest level evaluation system according to the third embodiment detects operational states including a scrolling velocity and a screen enlargement rate. Then, the user information managing server calculates interest levels for the parameter values based on the detected operational states and calculates amounts of change in the interest levels during the time intervals arranged in chronological order. Then, the user information managing server compares the change amounts and detects the same time interval during which change amounts calculated for two or more parameter values are equal to or larger than the predetermined threshold. Then, the user information managing server compares change amounts during a time interval preceding the detected same time interval and identifies a parameter value for which a change amount that is larger than a change amount calculated for the other parameter value during the preceding time interval has been calculated. Consequently, evaluation precision of the levels of users' interest in contents may be improved without depending on the contents.
• Next, a modified example of the embodiments is described.
• Although the first embodiment among the aforementioned embodiments exemplifies the case where the information processing terminal uses an operation time period and a scrolling velocity to calculate interest levels and amounts of change in the interest levels, the first embodiment is not limited to this. For example, the information processing terminal may use an operation time period, a no-operation time period, and a terminal momentum to calculate the interest levels. The information processing terminal may use an operation time period, a no-operation time period, a scrolling velocity, a screen enlargement rate, and a zoom operation time period to calculate the interest levels. In addition, the information processing terminal may execute a process of modifying a content based on the result of the process of evaluating levels of user's interest and execute a process of transmitting the results of the execution to the content server. By executing this, the efficiency of the process of modifying a content may be improved. In addition, in the second and third embodiments, the aforementioned modification process may be executed by the user information managing server.
• The second and third embodiments exemplify the case where each of the interest level evaluation systems is composed of an information processing terminal, the content server, and a user information managing server, but are not limited to this. For example, a proxy server may be installed in each of the interest level evaluation systems. Each of the information processing terminals acquires a content from the content server via the proxy server. If the proxy server is installed, the proxy server includes an analysis tag inserting section, an operational state storage section, an interest level quantifying section, and a temporal change comparing and detecting section, and each of the user information managing servers includes a user information storage section. In the case where each of the information processing terminals acquires a content from the content server, the analysis tag inserting section of the proxy server inserts a specific analysis tag in the content and enables the content to be analyzed. Based on the analysis tag inserted in the content, operational states of the users' information processing terminals are operation information, a terminal momentum, a screen enlargement rate, and the like and are acquired. The interest level quantifying section calculates interest levels for the parameter values based on the acquired operational states and calculates amounts of change in the interest levels. The change amounts for each of the parameter values are compared with the change amounts for the other parameter values, and the results of the comparison are stored in the user information storage section of the user information managing server.
• In each of the interest level evaluation systems according to the second and third embodiments, the analysis tag inserting section may be installed in each of the user information managing servers. In the case where each of the information processing terminals acquires a content from the content server, it is sufficient if each of the user information managing servers acquires the content via the analysis tag inserting section.
• Methods for calculating interest levels are not limited to the methods using the equations exemplified in the embodiments. For example, interest levels may be calculated according to the following Equation (4) obtained by combining the equations that are described in the embodiments and used to calculate interest levels.
• $\mathrm{Equation}\left(4\right)$$I\left(t\right)=\sum \frac{A\mathrm{time}\mathrm{period}\mathrm{for}a\mathrm{zoom}\mathrm{operation}}{\mathrm{The}\mathrm{terminal}\mathrm{momentum}\mathrm{weight}\mathrm{coefficient}w_d}+\sum \frac{A\mathrm{operation}\mathrm{time}\mathrm{period}}{\begin{array}{c}\mathrm{The}\mathrm{scrolling}\mathrm{velocity}\mathrm{weight}\mathrm{coefficient}w_{\mathrm{scr}}*\\ \mathrm{The}\mathrm{terminal}\mathrm{momentum}\mathrm{weight}\mathrm{coefficient}w_d\end{array}}*\mathrm{The}\mathrm{enlargement}\mathrm{rate}\mathrm{coefficient}z_{\mathrm{scr}}+\sum \frac{A\mathrm{no}\text{-}\mathrm{operation}\mathrm{time}\mathrm{period}}{\mathrm{The}\mathrm{terminal}\mathrm{momentum}\mathrm{weight}\mathrm{coefficient}w_{\mathrm{nop}}}*\mathrm{The}\mathrm{enlargement}\mathrm{rate}\mathrm{coefficient}z_{\mathrm{nop}}$
• In the case where interest levels are calculated using the aforementioned Equation (4), it is sufficient if the interest levels are calculated for each of combinations of parameter values indicating scrolling velocities and parameter values indicating terminal momentums.
• All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the invention and the concepts contributed by the inventor to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions, nor does the organization of such examples in the specification relate to a showing of the superiority and inferiority of the invention. Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

Claims (11)

What is claimed is:

1. An interest level evaluation method executed by a processor of a computer, the interest level evaluation method comprising:

detecting an operational state within a time period during which an information processing terminal displays a content;

calculating, for each of a plurality of parameter values related to the operational state, a level of interest in the content for each of a plurality of unit time periods based on the detected operational state and the plurality of parameter values;

calculating, for each of the plurality of parameter values, amounts of change in the level of interest during a plurality of time intervals arranged in chronological order based on the level of interest calculated for each of the plurality of unit time periods; and

evaluating the level of interest in the content by comparing the calculated amounts of change in the interest levels for each of the plurality of parameter values with the calculated amounts of change in the level of interest for another parameter value.

2. The interest level evaluation method according toclaim 1,

wherein the detecting includes

detecting a scroll operation as an operational state within a time period during which the information processing terminal displays the content,

wherein the interest level evaluation method further comprises

calculating a velocity of the scroll operation during an operation time period from a detection of the start of the scroll operation to a detection of termination of the scroll operation, and

wherein the plurality of parameter values include a plurality of values related to the velocity of the scroll operation.

3. The interest level evaluation method according toclaim 2, wherein the detecting further includes

detecting, as an operational state, a screen enlargement rate of the information processing terminal.

4. The interest level evaluation method according toclaim 1,

wherein the detecting includes detecting, as operational states, a time period for an operation of the information processing terminal and a momentum of the information processing terminal within the time period during which the information processing terminal displays the content, and

wherein the plurality of parameter values include a plurality of values related to the momentum of the information processing terminal.

5. The interest level evaluation method according toclaim 1, wherein the comparing includes

detecting a time interval during which change amounts calculated for the plurality of parameter values are equal to or larger than a predetermined threshold.

6. The interest level evaluation method according toclaim 5, wherein the comparing further includes:

comparing change amounts calculated for the plurality of parameter values during a time interval close to the time interval during which the change amounts are equal to or larger than the predetermined threshold, and

identifying a parameter value that is among the plurality of parameter values and for which a change amount that is larger than a change amount calculated for another parameter value has been calculated.

7. An interest level evaluating device comprising:

a memory; and

a processor coupled to the memory and configured to:

receive an input of a first scroll operation within a time period during which an information processing terminal displays a content,

calculating the velocity of the first scroll operation during a first operation time period from the reception of the input of the first scroll operation to the detection of the termination of the first scroll operation,

calculating the velocity of a second scroll operation during a second operation time period for the second scroll operation input and received after the first scroll operation, and

evaluating a level of interest in the content based on a change from the calculated velocity of the first scroll operation to the calculated velocity of the second scroll operation.

8. A computer-readable recording medium having stored therein a program for evaluating interest level of a content, the program execute a process comprising:

receiving an input of a first scroll operation within a time period during which an information processing terminal displays the content;

calculating a velocity of the first scroll operation during a first operation time period from a reception of the input first scroll operation to a detection of termination of the first scroll operation;

calculating a velocity of a second scroll operation during a second operation time period for the second scroll operation input and received after the first scroll operation; and

evaluating a level of interest in the content based on a change from the calculated velocity of the first scroll operation to the calculated velocity of the second scroll operation.

9. An interest level evaluation method executed by a processor of a computer, the interest level evaluation method comprising:

receiving an input of a first scroll operation within a time period during which an information processing terminal displays a content;

calculating a velocity of the first scroll operation during a first operation time period from a reception of the input first scroll operation to a detection of termination of the first scroll operation;

calculating a velocity of a second scroll operation during a second operation time period for the second scroll operation input and received after the first scroll operation; and

evaluating a level of interest in the content based on a change from the calculated velocity of the first scroll operation to the calculated velocity of the second scroll operation.

10. An interest level evaluation method executed by a processor of a computer, the interest level evaluation method comprising:

when an input of an operation within a time period during which an information processing terminal displays a content,

detecting an operation time from a reception of the input of the operation to a detection of termination of the operation, and

detecting a momentum of the information processing terminal within the time period during which the information processing terminal displays the content; and

evaluating a level of interest in the content based on the detected operation time and change of the motion.

11. The interest level evaluation method according toclaim 10, further comprising

detecting a screen enlargement rate of the information processing terminal within the time period during which the information processing terminal displays the content,

wherein the evaluating includes

evaluating the level of interest in the content based on a change in the detected screen enlargement rate and the change from the calculated velocity of the first scroll operation to the calculated velocity of the second scroll operation.

Images